158 / 2015-11-10 19:31:30

An improved adaptive pattern synthesis method based on SOCP and CS theory

In order to solve the problem existed in conventional adaptive pattern synthesis methods such as complex generalized inverse matrix process, complicated calculation process and a lot of memory space consumption. An adaptive pattern synthesis method based on second-order cone programming (SOCP) and compressed sensing theory is presented, which attempts to convert error performance function of Recursive Least Squares (RLS) algorithm into the form of standard second-order cone programming by mathematical transform so as to solve the above problem, and to use the compressed sensing theory in weights sparse processing so as to decrease the array elements and solving time. The simulation experiments show that the presented method is superior to traditional methods in pattern synthesis, and the obtained pattern has lowest side lobe level or deepest null steering.